devrel-content
Warn
Audited by Snyk on Mar 4, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The SKILL.md Phase 1 "Research & Validation" explicitly instructs the agent to "Google your topic" and "Search Reddit, HN, Stack Overflow" (public, user-generated sites) and to use those community signals to validate and shape the content, meaning untrusted third‑party content would be read and used to influence decisions and next actions.
Audit Metadata